It didn’t take anywhere near as long as I thought. I have two versions at the moment. One which doesn’t work and crashes a lot but has pretty graphics/labels and one which implements OpenCV stuff but is still using place holders. The crashes aren’t related, I started from scratch again today because the OpenCV libraries needed testing. So here’s a screen capture:
What’s happening here is that I can’t use the camera as the code is not signed and the iPhone Simulator is exactly that. It’s not an emulator and thus cannot do all sorts of things, like say use a Mac’s built in camera. So we select an existing photo (which I downloaded into the simulator using safari and browsing the local file system), what happens next is that the subsequent page does all of the processing whilst it loads the view. This explains the minor delay. What is also happening here so I can test things is that the initial image is displayed in the view but also added as the first frame of an animation. Each subsequent processing step is added as another frame. When the view is loaded the place holder button in the top right fires off a function that plays the animation over 5 seconds so I can get an overview of the image processing before progressing the application. Simple?
It’s been frustrating but I really don’t think it’s bad to go from buying a Mac on Wednesday to having working iPhone code using external C++ based libraries in 3 days. That includes learning Objective C, the iPhone API and a bit more of XCode although I have used it before. Also gdb, which I ♥. If I had to thank all the people who wrote tutorials and made video examples this would be a rather long article but I’ve thanked people on their blogs where they’ve given me major insight.